C and Cl Isotope Fractionation of 1,2-Dichloroethane Displays Unique

Jul 10, 2014 - ... 1.0070 ± 0.0002 (13C-AKIE, oxidation), 1.068 ± 0.001 (13C-AKIE, SN2), and 1.0087 ± 0.0002 (37Cl-AKIE, SN2) fell within expected ...
0 downloads 0 Views 792KB Size
Article pubs.acs.org/est

C and Cl Isotope Fractionation of 1,2-Dichloroethane Displays Unique δ13C/δ37Cl Patterns for Pathway Identification and Reveals Surprising C−Cl Bond Involvement in Microbial Oxidation Jordi Palau,*,† Stefan Cretnik,‡ Orfan Shouakar-Stash,§ Martina Höche,‡ Martin Elsner,‡ and Daniel Hunkeler† †

Centre for Hydrogeology and Geothermics, University of Neuchâtel, 2000 Neuchâtel, Switzerland Helmholtz Zentrum München, German Research Center for Environmental Health, D-85764 Neuherberg, Germany § Department of Earth and Environmental Sciences, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada ‡

S Supporting Information *

ABSTRACT: This study investigates dual element isotope fractionation during aerobic biodegradation of 1,2-dichloroethane (1,2-DCA) via oxidative cleavage of a C−H bond (Pseudomonas sp. strain DCA1) versus C−Cl bond cleavage by SN2 reaction (Xanthobacter autotrophicus GJ10 and Ancylobacter aquaticus AD20). Compound-specific chlorine isotope analysis of 1,2-DCA was performed for the first time, and isotope fractionation (εCl bulk) was determined by measurements of the same samples in three different laboratories using two gas chromatography−isotope ratio mass spectrometry systems and one gas chromatography−quadrupole mass spectrometry system. Strongly pathway-dependent slopes (Δδ13C/Δδ37Cl), 0.78 ± 0.03 (oxidation) and 7.7 ± 0.2 (SN2), delineate the potential of the dual isotope approach to identify 1,2-DCA degradation pathways in the field. In contrast to different εCbulk values [−3.5 ± 0.1‰ (oxidation) and −31.9 ± 0.7 and −32.0 ± 0.9‰ (SN2)], the obtained εCl bulk values were surprisingly similar for the two pathways: −3.8 ± 0.2‰ (oxidation) and −4.2 ± 0.1 and −4.4 ± 0.2‰ (SN2). Apparent kinetic isotope effects (AKIEs) of 1.0070 ± 0.0002 (13C-AKIE, oxidation), 1.068 ± 0.001 (13C-AKIE, SN2), and 1.0087 ± 0.0002 (37ClAKIE, SN2) fell within expected ranges. In contrast, an unexpectedly large secondary 37Cl-AKIE of 1.0038 ± 0.0002 reveals a hitherto unrecognized involvement of C−Cl bonds in microbial C−H bond oxidation. Our two-dimensional isotope fractionation patterns allow for the first time reliable 1,2-DCA degradation pathway identification in the field, which unlocks the full potential of isotope applications for this important groundwater contaminant.



INTRODUCTION

Scheme 1

Chlorinated ethanes are among the most widespread contaminants in groundwater,1 and 1,2-dichloroethane (1,2DCA) has been found in 36% of 1585 National Priorities List sites identified by the U.S. Environmental Protection Agency.2 The presence of 1,2-DCA, an intermediate in the production of plastics, in groundwater is mainly a consequence of industrial activity.2 A number of laboratory3−9 and field10,11 studies showed 1,2-DCA biodegradation under aerobic3−6 and anaerobic4,7−11 conditions via different reaction pathways.9 Under aerobic conditions, 1,2-DCA can be degraded either via nucleophilic substitution (SN2)5,6 or via oxidative cleavage of a C−H bond3 catalyzed by a hydrolytic dehalogenase or monooxygenase enzyme, respectively (Scheme 1). Initial products of both reactions, 2-chloroethanol (SN2 reaction) and 1,2-dichloroethanol (oxidation), are further degraded to ubiquitous end products, which hampers a direct identification of degradation pathways from metabolite analysis. Alternative approaches to detecting and identifying 1,2-DCA transformation pathways in the subsurface are therefore warranted. © 2014 American Chemical Society

This is crucial information in environmental field studies for assessing 1,2-DCA natural attenuation. Compound-specific isotope analysis (CSIA) is an innovative tool for investigating degradation pathways of organic Received: July 1, 2014 Accepted: July 10, 2014 Published: July 10, 2014 9430

dx.doi.org/10.1021/es5031917 | Environ. Sci. Technol. 2014, 48, 9430−9437

Environmental Science & Technology

Article

contaminants.12−14 Isotope ratios of individual compounds, measured either by gas chromatography and isotope ratio mass spectrometry (GC−IRMS) or gas chromatography and quadrupole mass spectrometry (GC−qMS), are reported using the delta notation: h

δ E sample =

R(hE/1E)sample R(hE/1E)standard

chlorine leaving group KIE would be expected, whereas in the oxidation reaction, chlorine atoms sit next to the reacting bond so that only a secondary KIE would be expected (Scheme 1). Secondary isotope effects at positions next to the reacting bond are generally much smaller than primary isotope effects.12 Until recently, Cl-CSIA of chlorinated aliphatic compounds was not feasible, however, because a direct method that would produce a suitable Cl-containing measurement gas inside a chromatographic separation gas was lacking. However, new analytical methods were developed on the basis of the measurement of selected isotopologue ions or isotopologue ion fragments in unconverted analyte molecules using both continuous flow GC−IRMS23 and GC−qMS.24−26 In addition, a theoretical framework provided the theoretical justification for such evaluation of isotope fractionation from ion current ratios of molecular and fragment ion multiplets. 27 A recent interlaboratory study took the next step and demonstrated that comparable δ37Cl values were obtained upon analysis of a set of pure trichlorethene (TCE) standards on eight different instruments.28 Because the technology is so new, however, a comparative study that shows that comparable εCl values are obtained when analyzing degradation samples on different instruments and in different laboratories would also be desired. No such study has yet been conducted. Most notably, Cl-CSIA studies have so far been applied to only few compounds, because two isotopically distinct compound-specific standards are necessary for every new substance.28 This has restricted applications primarily to chlorinated ethylenes,18,19,29−34 so that, to the best of our knowledge, dual element isotope data are currently nonexistent for chlorinated ethanes. This study, therefore, aimed (i) to establish for the first time dual element (C and Cl) isotope analysis of the chlorinated ethane 1,2-DCA, (ii) to perform the Cl isotopic analysis in three different laboratories (i.e., Waterloo, München, and Neuchâtel), using two different GC−IRMS systems and one GC−qMS system, to investigate the consistency of εCl values obtained with different instruments and analytical methods, and (iii) to investigate carbon and chlorine isotope fractionation during aerobic biodegradation of 1,2-DCA with three pure strains to determine whether the dual isotope slopes are sufficiently different to potentially distinguish between hydrolytic dehalogenation (SN2) and oxidation (C−H bond cleavage) in the field.

−1 (1)

where R is the isotope ratio of heavy (hE) and light (lE) isotopes of element E (e.g., 13C/12C and 37Cl/35Cl). The isotope fractionation (ε) expresses by how much hE/lE is smaller (negative values) or larger (positive values) in the average of freshly formed products compared to that in the substrate from which they are produced. Transformationinduced isotope fractionation is generally larger than the one related to phase transfer processes such as sorption or volatilization.15 Bulk (i.e., compound-average) ε values can be calculated using a modified form of the Rayleigh distillation equation: ln

⎛ δ hE + 1 ⎞ Rt ⎟⎟ = εbulk × ln f = ln⎜⎜ h t R0 ⎝ δ E0 + 1 ⎠

(2)

where Rt and R0 are the current and initial isotope ratios, respectively, and f is the remaining fraction of the compound. Previous laboratory experiments16 showed that for 1,2-DCA different carbon εbulk values of −29.2 and −3.9‰ reflected different reaction pathways: hydrolytic dehalogenation (C−Cl bond cleavage via SN2) and oxidation (C−H bond cleavage), respectively (Scheme 1). Knowledge of in situ contaminant biodegradation reactions is essential for evaluating the fate and long-term impact of 1,2-DCA in groundwater. For aerobic biodegradation of 1,2-DCA, isotope data are particularly important as no characteristic products accumulate. However, while isotope fractionation of one element alone can distinguish pathways in laboratory experiments (where mass balances can be established and εbulk values determined), this is not possible under field conditions. Here, evidence of a second element and a dual isotope approach is necessary to distinguish 1,2-DCA degradation pathways.17 As observed experimentally,18,19 for a given compound, combined isotope analysis of two elements (e.g., δ13C vs δ37Cl) during the course of a reaction generally yields a linear trend in a dual element isotope plot with a slope characteristic of the reaction mechanism. The reason is that the dual element isotope slope (Λ = Δδ13C/Δδ37Cl, where Δδ13C and Δδ37Cl are changes in isotope ratios during degradation) reflects isotope fractionation of both elements.12 Therefore, different slopes are expected from distinct reaction pathways involving different bonds with different elements. Currently, there is growing interest in dual element isotope analysis for improved differentiation of transformation mechanisms, and several authors pointed out that complementary mechanistic insight for 1,2-DCA aerobic biodegradation reactions could be achieved by the additional analysis of chlorine12 and/or hydrogen16,20 isotope ratios. The reason is that isotope fractionation can be traced back to underlying kinetic isotope effects, which are highly reaction-specific.21 During enzymatic degradation, molecules containing the light isotope at the reactive site (e.g., 35Cl) typically exhibit a reaction rate (e.g., 35k) higher than the rate of those with a heavy isotope (e.g., 37Cl), resulting in a kinetic isotope effect (KIE) of 35k/37k.22 When a C−Cl bond is broken, a (primary)



MATERIALS AND METHODS Preparation of Pure Cultures. Three pure strains with known initial biotransformation mechanisms were used for the batch experiments: Pseudomonas sp. strain DCA1 (oxidation) and Xanthobacter autotrophicus GJ10 and Ancylobacter aquaticus AD20 (SN2 reaction). X. autotrophicus GJ10 (DSMZ 3874) and A. aquaticus AD20 (DSMZ 9000) were purchased (DSMZ, Braunschweig, Germany), and Pseudomonas sp. strain DCA1 was kindly provided by E. Edwards (Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON). The growth medium was prepared as described by Hunkeler and Aravena.35 Cultures and experiments were prepared in 250 mL glass bottles, which contained 185 mL of medium and were capped with Mininert-Valves (Vici Precision Sampling, Baton Rouge, LA). Cultures were amended with 1,2-DCA (Fluka, ≥99.5% pure) and incubated in the dark at room temperature while being continuous shaken (100 rpm). Headspace 1,2-DCA concentrations were monitored throughout the incubation 9431

dx.doi.org/10.1021/es5031917 | Environ. Sci. Technol. 2014, 48, 9430−9437

Environmental Science & Technology

Article

peaks are much higher than those of the parent ion peaks. Both ion couples (m/z 100 and 102 and m/z 62 and 64) correspond to isotopologue pairs ([37Cl212C21H4]+ and [37Cl35Cl12C21H4]+, respectively, and [37Cl12C21H3]+ and [35Cl12C21H3]+, respectively) that differ by one heavy chlorine isotope. The isotope ratio (R) can be obtained from the ratio of these isotopologues according to eq 327

period. Before the experiments were started, the cultures were transferred three times. Each subculture was spiked with pure 1,2-DCA four times before a 15 mL aliquot was transferred to 170 mL of autoclaved fresh medium. The spike volume was 9 μL of pure 1,2-DCA for the first and second subcultures and 22.5 μL for the third, leading to aqueous phase concentrations of 0.6 and 1.5 mM, respectively. Experiment Sampling. All experiments were conducted in triplicate. Experiments and controls were amended with 22.5 μL of pure 1,2-DCA, corresponding to the chlorine isotopic working standard (δ37Cl0-CHYN2 = 0.8 ± 0.1‰), to produce an initial aqueous concentration of 1.5 mM. Bottles were shaken upside down to prevent leakage of the gas phase through the valve. After being shaken for 1 h, samples representing the initial concentration were collected. For concentration and isotopic analysis, aqueous samples (1.5 mL) were taken at selected time points and preserved frozen36 in 2 mL vials with NaN3 (1 g/L). Two abiotic control bottles were prepared with 185 mL of autoclaved mineral medium, and samples were collected and preserved as described for the experiments. For each culture experiment, triplicate samples were shipped frozen to the University of Waterloo and to the Helmholtz Zentrum München for chlorine isotope measurements. Isotopic and Concentration Analysis. Five pure 1,2DCA isotopic working standards, one for carbon and four for chlorine, were used for instrument monitoring and external calibration of sample raw δ37Cl values to the international Standard Mean Ocean Chloride (SMOC) scale. The isotopic signature of the carbon standard (δ13CV‑PDB = −29.47 ± 0.05‰) was determined beforehand using an elemental analyzer coupled to an IRMS system. With regard to the chlorine standards, CHYN1 and CHYN2 (δ37ClSMOC values of 6.30 ± 0.06 and 0.84 ± 0.14‰, respectively) were characterized relative to SMOC in München by IRMS after conversion of 1,2-DCA to methyl chloride according to the method of Holt et al.37 IT2-3001 and IT2-3002 (δ37ClSMOC values of 0.83 ± 0.09 and −0.19 ± 0.12‰, respectively) were calibrated against the CHYN standards using a GC−IRMS system in Waterloo and a GC−qMS system in Neuchâtel, respectively. A detailed description of analytical methods is available in the Supporting Information. Carbon isotope ratios were measured by GC−IRMS, and the precision based on the working standard δ13C value reproducibility was 0.5‰ (1σ). Chlorine isotopic analysis was performed using the following instruments: (1) in Waterloo, a model 6890 GC instrument (Agilent, Santa Clara, CA) coupled to a continuous flow IsoPrime IRMS instrument (Micromass, Manchester, U.K.; currently Isoprime Ltd.); (2) in München, a TRACE GC instrument (Thermo Fisher Scientific, Milan, Italy) directly coupled to a Finnigan MAT 253 IRMS instrument (Thermo Fisher Scientific, Bremen, Germany); and (3) in Neuchâtel, a model 7890A GC instrument coupled to a model 5975C quadrupole MS instrument (Agilent). The instrument used in Neuchâtel will be termed the qMS instrument, and those used in Waterloo and München will be termed IRMS-1 and IRMS-2 instruments, respectively. For analyzing 1,2-DCA, two ions of the molecular group (m/ z 100 and 102) were measured with the IRMS-2 instrument, whereas the two most abundant fragment ions (m/z 62 and 64) were used for the IRMS-1 instrument and the qMS instrument. For 1,2-DCA, the intensities of the most abundant fragment ion

37

37

R=

p

Cl = 35 Cl

=2×

35

p

=

102

64

100

62

I = I

I I

37 Cl(k)35Cl(n − k) k × 37 n−k+1 Cl(k − 1)35Cl(n − k + 1)

(3)

where 37p and 35p are the probabilities of encountering 37Cl and 35 Cl, respectively, n is the number of Cl atoms, k is the number of 37 Cl isotopes, 37 Cl (k) 35 Cl (n−k) and 37 Cl (k−1) 35 Cl (n−k+1) represent the isotopologues containing k and k − 1 heavy isotopes, respectively, and I indicates the peak intensity of each ion. For the qMS instrument, isotope ratios were calculated using eq 3 and raw δ37Cl values were determined by referencing versus an external 1,2-DCA working standard according to eq 1. In this case, the standard was dissolved in water and measured like the samples in the same sequence.26 In the IRMS-1 and IRMS-2 instruments, raw δ37Cl values were determined by automatic evaluation of sample’s target ion peaks against the ion peaks of the 1,2-DCA monitoring gas, which was introduced by a dual inlet system during each sample run providing an anchor between sample measurements.28 Subsequently, a two-point linear calibration of raw δ37Cl values to SMOC scale was performed in each laboratory using two external working standards, i.e., IT2(3001 and 3002) for the IRMS-1 instrument and CHYN(1 and 2) for the IRMS-2 and qMS instruments.28 The analysis schemes applied in each laboratory are available as Supporting Information. For the measurements with the qMS instrument, samples and standards were diluted to a similar concentration and each of them was measured 10 times, leading to a precision (1σ) on the analysis of standards of ±0.3‰. For the IRMS-1 and IRMS-2 instruments, the precision on the analysis of standards was ±0.1‰ (1σ). The δ37Cl of controls remained constant (0.7 ± 0.2‰; n = 12) at δ37Cl0 within the analytical uncertainty throughout the experiment. Concentrations of 1,2-DCA were measured by headspace analysis using a TRACE GC-DSQII MS instrument (Thermo Fisher Scientific, Waltham, MA) in single-ion mode (m/z 51, 62, 64, 98, and 100). Concentrations were determined using a five-point calibration curve. The estimated total relative error on analysis of external standards interspersed along the sequence was ±4%. Concentrations were corrected for 1,2DCA volatilization to the bottle headspace using Henry’s law and Henry coefficients.38 An aqueous phase 1,2-DCA concentration decrease of